Posts Tagged ‘book indexes’

Index Formats

Monday, May 7th, 2018

Run-In Versus Indented Indexes

This is the reprint of a blog posting that was originally printed on September 27, 2010.

The Chicago Manual of Style describes two index formats that publishers use: run-in and indented.  In run-in style, the subentries follow one another without each one starting a new line.  In indented style, each subentry begins a new line and is indented (usually one em).  The Chicago Manual of Style gives further specifics for each case.

The Chicago Manual of Style and many other scholarly publishers generally prefer run-in style because it requires less space.  It works best, however, when there is only one level of subhead.  However, the book itself, The Chicago Manual of Style, has an index in indented style.  Indented style is usually preferred in scientific works and reference works.  It is particularly useful where sub-subentries are required.

Each publishing house usually has its own indexing style guidelines, specifying which index format to use for its books.  This format is selected based on space considerations or on the material, since a work with a lot of scientific terminology would be easier to read in indented format.

For more information on run-in or indented  indexes, refer to The Chicago Manual of Style.  For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

 

Solutions for the Index Locator Problem with e-Book Indexes

Wednesday, August 29th, 2012

A paragraph-numbering system for locators, such as the one the Chicago Manual of Style uses, might possibly work for linking all e-book version of indexes the same way.  But this might happen only if all e-reader devices would use a paragraph-based system, a standardized method of numbering screen pages/locations, and consistent placement of the index anchor links on their screen “pages.”

If we want both quick and precise post-coordinate indexing for finding information in e-books, we have to find a way to combine machine-based indexing with the kind of pre-coordinate indexing that indexers do for print books even if whole indexes in e-books are not visible to the reader.  We need a semi-pre-coordinated, post-coordinate system that can sift through a large amount of data while at the same time be able to find a relevant, precise, piece of data.

This kind of indexing is “latticed” indexing.  We need to develop a “latticed” indexing system for sifting through large quantities of data and at the same time connecting a lot of related bits of data together in an organized way.  The use of the flat, one-dimensional alphabetical index, separate from the text in the print book, is both tedious to implement and antiquated in dynamic digital space.  We need a multi-dimensional system of information storage and retrieval that’s appropriate for cyberspace.  A place to begin to envision what a “latticed” index on the Web might look like would be a “faceted” classification system, and would offer e-book readers a way to search more effectively.

For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

Still More Problems with e-Book Indexes

Friday, July 6th, 2012

The E-Book Screen Locator Problem

Linked indexes are vital in an e-book because a reader has to “flip” through hundreds of screens of the e-book one at a time to get to a particular page.  The way each type of e-reader divides up an e-book also impacts the text reflow issue for indexes.  Kindle books and iBooks don’t have the same number of “pages,” and, in fact, don’t even use the same word.  While Apple calls its pages “pages,” Kindle calls its pages “locations.”  Amazon numbers the paragraphs in a book and calculates a percentage of the total e-book text to define what the number is for each particular “location” in a Kindle e-book.  Apple uses a different way to create virtual pages and to mark the beginning and end boundaries for an e-book file and to apportion the number of “pages” in its iBooks.

Despite the varying number of either “pages” or “locations” in an e-book, indexers need to be able to drop an anchor into either kind of e-book in order to link a page number in the index to the relevant discussion within the digital text.  In either kind of e-book the hyperlinked print-book page number should be able to take the reader to the right “page” or “location” in the e-book text to begin exploring the topic.

I will offer some suggestions concerning what to do about the index locator problem in the next blog posting.  For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

 

More Problems with e-Book Indexes

Wednesday, June 6th, 2012

The E-Reader Device Problem

To distinguish itself from others and to establish its own unique competitive advantage, each e-reader manufacturer uses different methods for displaying an e-book.  These differences include the various size of the physical “frame” for the screen display in the book, as well as the types of navigation methods used, including icons and features that vary, and even the different ways of adding virtual page numbers for each e-reader.

As a result, this variation affects index locators.  Index locators are linked back to the text with a device called an “anchor,” which is a piece of code that is invisible to the reader.  A “page” anchor on an e-book screen corresponds to the beginning of a page in the print version of a book.

Amazon Kindle places anchors in the upper left hand of the e-book screen, while Apple iPad places them in the middle.  Because the iPad has a larger screen, users can rotate the iPad 90 degrees and a two-page image will appear on the screen.  Readers may find the anchor in the middle of the left-hand “page” or in the middle of the right-hand “page” on the iPad screen, and may have to read two e-book pages before finding the beginning of an indexed discussion.

In addition, text in e-books can reflow over a number of pages depending on how large the font is made by the book’s reader.  Consequently, page number ranges in an e-book index are useful only to a very limited extent.  A page range of three pages could become six or even more e-book “pages” if the font size of the e-reader device is changed.

I will discuss another problem with e-book indexes in a future blog posting.  For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

Problems with e-Book Indexes

Thursday, May 3rd, 2012

E-Book Text-Reflow Problem

Why do the vast majority of e-books today come without indexes?  Primarily, because of implementation challenges.  A reader’s ability to adjust font size plays havoc with page references.  Where, for example, in an e-book should index entries point to?  At certain font sizes, hyperlinked numbers might lead exactly where you want to go.  Or you might have to page ahead – or back – a click or three to find it.  Which direction?  Varying screen sizes will compound matters.

Indexers are familiar with page reflow. That problem is why we warn our clients to make sure the pages they submit for an index are in FINAL form.  This problem is magnified ten times over in e-books.  The reader of an e-book can adjust their font size.  Some readers of e-books like big font; some prefer smaller font.  Now all have many choices of font-size when they use an e-reader device.

Published e-books have a serious text reflow problem.  This problem makes it almost impossible to include hyperlinks to page references that direct the reader to exactly where you want them to go.

I will discuss other problems with e-book indexes in future blog postings.  For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

 

Missing the e-Book Index

Wednesday, April 11th, 2012

A well-designed index can help facilitate smarter e-books, making it easier for readers to find and retrieve information more efficiently.  Unfortunately, at least for the immediate future, indexes won’t likely be appearing in most e-books on the the main e-reading systems, such as Kindle and Nook.  The companies behind these reading platforms haven’t programmed their reading software in a way that would make indexes easy – or at least easier – to implement.  Amazon comes closest, with the page numbers you now see on some Kindle books appearing courtesy of so-called “page list” data that many publishers provide – essentially lists correlating e-book location with pages in a print book.  That data could, in theory, be used to automate the implementation of indexes and to make them more accurate.  The indexer needs to move beyond taking a digital snapshot of a print index and turning each entry into a hyperlink.  Instead, search tools and indexes need to work in partnership, passing queries from one to the other as different readers arrive with different questions.

In following blog postings, I will explore specific problems with e-book indexes.  For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

 

Long Strings of Reference Locators in an Index

Thursday, December 23rd, 2010

As a follow-up to my last posting, Guidelines for Editing Indexes, I wanted to explore in detail one of the guidelines.  A main entry followed by a long string of undifferentiated reference locators or page numbers at the main heading level should be broken down into subheadings.  A common rule of thumb is that more than five or six reference locators should be differentiated by the addition of subheadings.

Leaving a long string of page numbers will force the reader to plow through many pages before finding the desired information.  There might be 10 or 15 page numbers listed, a daunting task for the reader to wade through in search of the one elusive piece of information.  Exactly what constitutes a “long” string of page numbers is often defined in a publisher’s style guide or by the space allowed for the index.  The reason for adding subheadings is simple.  Subheadings subdivide and characterize the  information and make it easy to find the information in question.

For more information, see Chapter Nine, “Editing the Index,” of Nancy C. Mulvancy’s Indexing Books, Second Edition, Chicago: University of Chicago Press, 2005, pp.219-241.  For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

Guidelines for Editing Indexes

Thursday, November 18th, 2010

When the indexer finishes compiling the last line of an index, it is only the first draft, and like any piece of writing, it requires editing.  Substantive editing tasks, rewriting or reorganization of material, should be undertaken by the indexer alone.  The indexer should produce an index manuscript that is in need of few copyediting or proofreading changes.

Nancy C. Mulvany gives an excellent discussion of editing indexes in Chapter Nine, “Editing the Index,” of Indexing Books, Second Edition, Chicago: University of Chicago Press, 2005, pp.219-241.  She walks the indexer though the substantive editing tasks that he or she must undertake and gives guidelines to follow.

  • Main headings.  Evaluate one-by-one.  Are they clear and concise?  Do they make sense?  Is the wording evident so that a reader may look them up?
  • Long strings of undifferentiated reference locators at the main heading level should be broken down into subheadings.  More than 5 or 6 reference locators should be differentiated by the addition of subheadings.
  • Subheadings.  Evaluate for clarity and conciseness.  Do they make sense?  Is the wording evident?  Are they necessary?
  • A single subheading under a main entry and the group of subheadings that all have the same page number should be condensed.  Subheadings that all have the same page number should be eliminated or converted to main entries.
  • Cross-references and double-postings.  Each cross-reference in an index must be verified and must be direct.  All circular cross-references must be eliminated.  Cross-references that send readers elsewhere in the index only to pick up one or two references can often be changed to double-postings, where the same information is entered in two places.

Copyediting and proofreading tasks for the indexer include alphabetizing, spelling, parallel construction, punctuation, and capitalization.  For more information on editing indexes, see Mulvany’s book, Indexing Books.  For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

Indexes in E-Books

Friday, April 16th, 2010

Electronic book readers such as the Kindle and the Nook have changed the way people read books. With the electronic book reader, pages are turned electronically and silently, but what happens when the user tries to look up a subject in the index?  The index is static.  It has the same list of subjects and page references as the print version, but requires that the user tediously page through the alphabet to find the subject, then wade through the text to find the referenced page.

The publishing industry has a great opportunity to make indexes dynamic in e-books by adding hyperlinks to index references.  By clicking on an index reference, the reader would be linked to the appropriate section in the book.  Embedded indexing offers this technology, by creating index entries that are inserted electronically into the computer files along with the text of the document.  Instead of creating a separate index file with dedicated indexing software, an embedded index is created in the same software as the rest of the document.

Currently, print documents produced with embedded indexing can be posted online.  When the user clicks on the index entry, the relevant text is displayed because index references have been turned into links.  Adding hyperlinks to index references in e-book indexes would be the next step.

For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com

Concordances vs. Indexes

Saturday, March 6th, 2010

An index requires a degree of sophistication and skill lacking in a computer-generated concordance.  A concordance is an alphabetical list of words with locations in the text, a somewhat rudimentary index.  The computer automates the process, parsing the text and collecting the index words and the locations in the text, then writing the generated concordance to a PDF or text file.  The computer greatly reduces your time and effort, but the results are far inferior, for a number of reasons.

A computer cannot distinguish between what is important and what is not.  Many concepts that are mentioned only in passing are not useful to the reader, but are included along with all other concepts.  Readers are quickly overwhelmed.  In addition, a concordance does not cross-reference subjects.  A concordance may also leave out a subject that is discussed at length without ever being mentioned by name.

Indexers can adapt and stay true to style guidelines, while a computer-generated concordance would not.  While an indexer is self-reflective, visiting and revisiting the index to improve upon it, a computer merely generates a list of words contained within the text, then the work stops there.

Computers are capable of many things, but creating a high-quality index is not one of them.

For more information about the services provided by the author of this blog, see the Stellar Searches LLC website, http://www.stellarsearches.com